The role of diet in multiple sclerosis

Multiple sclerosis (MS) is an inflammatory autoimmune/neurodegenerative disorder of the central nervous system that often manifests itself at a crucial time in a person’s career and family development. Australian dietary recommendations for those with MS only extend to adherence to the Australian Guidelines for Healthy Eating as, currently, there is limited evidence whether and how diet influences MS progression or inflammatory activity. Evidence from previous epidemiological research shows that most dietary assessments were conducted by focusing on single foods and nutrients. Clinical trial results were often inconclusive, with some finding associations while others were unable to confirm findings. Bias in interventions is also common. Limited research has focused on diet composite scores, which combine multiple food groups, individual foods, and/or nutrients into a score based on a specific dietary theoretical construct. Composite diet scores may be helpful to understand the complexities of diet and how individual nutrient associations fit together. Therefore, the primary aim of this thesis is to conduct studies on composite diet scores and indices derived by the Food Frequency Questionnaire that captured the quality (Diet Quality Tracker and Australian Recommended Food Score), inflammatory potential (Dietary Inflammatory Index), and acidity of food intake (Potential Renal Acid Load and Net Endogenous Acid production). Associations between these diet scores and various MS-related symptoms were examined to generate novel information which could inform the selection of trial study endpoints and duration of trials. This information could also assist with the development of diet interventions that could then be tested in randomised controlled trials.

This thesis used data from the Ausimmune Longitudinal (AusLong) Study. The AusLong Study is an ongoing prospective study since 2003 that is a continuation of the original cross-sectional Ausimmune case-control study. The Ausimmune study is a multicentre case-control study including people recruited from four regions of Australia (Queensland, New South Wales, Victoria, and Tasmania) who had a first clinical diagnosis of CNS demyelination (n=282). The first study of this thesis examined the association between diet quality and depression, anxiety, and fatigue in MS. This study found some evidence that a higher-quality diet, at least with the Australian Recommended Food Score, was associated with a lower level of depression and anxiety. There was no convincing evidence that diet quality was associated with levels of fatigue, however. The second study, focused on identifying associations between diet quality and hazard of relapse, annualised change in disability as measured by the Expanded Disability Status Scale and Magnetic Resonance Imaging (MRI) lesion volume in MS. This study found evidence that a higher-quality diet as measured by the Australian Recommended Food Score, was associated with lower inflammatory disease activity as assessed by MRI periventricular fluid-attenuated inversion recovery (FLAIR) lesion volume. Less evidence was found with other measures of inflammatory activity or progression; any associations were lacking in significance or dose dependency. As a collective, these two studies provide some evidence that diet quality is associated with measures of MS progression or inflammatory activity. There was inconsistency between the two measures used, by which the ARFS was consistently associated but the DQT wasn’t. The effects were often modest, a dietary measure that correlates with diet quality could explain our observed associations.

The third study of this thesis examined the association between the Dietary Inflammatory Index and depression, anxiety, and fatigue in MS. A higher Dietary Inflammatory Index was associated with higher levels of depression and anxiety five years later and a greater subsequent worsening in depression and anxiety scores over five years. In addition, cumulative residual energy-adjusted Dietary Inflammatory Index scores were associated with depression and anxiety at the 10-year review. There were no cross-sectional associations between the Dietary Inflammatory Index scores with depression and anxiety. No convincing associations were observed with fatigue in either cross-sectional or prospective analyses.

Associations between the dietary inflammatory index and risk of relapse, annualised Expanded Disability Status Scale, and MRI lesion volume in MS were examined in the fourth study of this thesis. This showed that a higher Dietary Inflammatory Index was prospectively associated with a higher hazard of relapse and a higher FLAIR lesion volume overall, as well as in the periventricular region. No associations were observed with annualised Expanded Disability Status Scale change or black hole lesion volume. As a collective, these studies suggests that a proinflammatory diet is associated with a worsening of some MS outcomes.

The fifth study focused on examining associations between dietary acid load and depression, anxiety, and fatigue in MS. A higher dietary acid load (Potential Renal Acid Load and Net Endogenous Acid production) was associated with an increase in absolute value and change in depression levels over five years. Level of depression at the 10-year review was predicted by both the baseline dietary acid scores as well as the changes in acid scores from baseline to 5-year and from 5- to 10-year reviews. Some associations in the same direction were observed with anxiety and fatigue but were less consistent. The sixth study found no association between dietary acid load and rate of relapse, annualised Expanded Disability Status Scale change, and MRI lesion volume in MS.

As associations were found for more than one composite score with the same outcome, for example, both PRAL and DII were associated with depression, we examined the correlations between the dietary composite scores. Mutually adjusted multivariable analyses were also conducted to determine which composite dietary score was most strongly associated. Two comparison analyses were performed. When including all composite dietary predictors within the same model to determine which composite diet score was most strongly associated with depression and anxiety, a higher dietary acid load score was most strongly associated. When standardised regression coefficients were used to compare the effects of each composite diet score on the FLAIR lesion volume, a higher dietary inflammatory index appeared to be most strongly associated with an increase in subsequent FLAIR lesion volume.

In summary, data from this thesis has provided novel insights into the role of diet in MS progression and inflammatory outcomes over a 10-year disease period starting at the earliest clinical manifestation of MS and support the utility of derived composite scores as being useful in MS diet research. The findings in this thesis require replication before proceeding to clinical trials to test whether diets that are low in dietary inflammatory potential or dietary acid load have beneficial outcomes in MS. The findings, together with other observational studies, will be useful to guide the design choices for randomised controlled trials of diet in MS, including the outcome measures, duration of the trial, and features of the diet.